Project 1. Summary/Abstract Improving macromolecular crystallization using lattice-promoting variants of GFP The bottleneck in structure determination by X-ray crystallography is crystallization, where -70% of purified proteins fail. A major reason proteins fail to crystallize is insufficient suitable lattice-forming contacts. The central hypothesis in this project is that the probability of obtaining crystals is increased every time a new form of a molecule with new potential crystal contacts is tested. We will develop rapid and general methods to generate numerous venations on a protein molecule by linking the protein to a series of oligomeric forms of green fluorescent protein (GFP). The use of GFP is a major advantage because it can be linked firmly to practically any protein by inserting a hairpin (two strands connected by a loop) from GFP into a loop of the target protein. The target protein containing this extra hairpin can bind tightly to a